AI-Integrated Structural Optimization Framework for Lightweight Heavy Fabrication Systems in Smart Manufacturing Environments

Abstract

Heavy fabrication systems, such as those used in the mining, construction, off-highway vehicle, and industrial equipment industries, are usually developed through conservative engineering practices, resulting in overdesigns in terms of structure, leading to excessive material usage. Although such practices ensure the durability and safety of the structure, they also result in increased manufacturing costs, energy consumption, and carbon emissions. The existing optimization techniques, such as topology optimization and size optimization, may not consider the manufacturability constraints of the welded structure, as well as the lack of consideration of real-world operational feedback. This paper proposes a new framework for AI-assisted structural optimization for lightweight heavy fabrication systems in a smart manufacturing paradigm. This new optimization framework includes physics-based finite element methods, machine learning-based surrogate modeling, multi-objective optimization, manufacturability-based constraint embedding, and digital twin-based feedback control in a unified architecture. The optimization process targets the minimum weight design subject to static strength, vibration, fatigue, and weld-related design constraints, while manufacturability is ensured through thickness limits, weld accessibility, and assembly feasibility. The incorporation of operational sensor information aids in the continuous improvement of predictive models, thereby allowing structural optimization to become a dynamic process rather than a static activity limited to the design phase. The proposed method improves design iterations, reduces material usage, improves vibration response, and meets sustainability goals by incorporating embedded carbon impact modeling. This proposed method moves heavy fabrication engineering into an adaptive, intelligence-based, and sustainability-focused structural system that meets Industry 4.0 and Industry 5.0 concepts.